Psychology in the 21st C – Getting over our addiction to p so our research can be evidence for our practice.
Psychology in the 21st C – Getting over our addiction to p so our research can be evidence for our practice Neville M Blampied University of Canterbury
In the middle years of the 20th C two things happened that had far-reaching impacts on psychology. The first was the invention by R.A Fisher and other statisticians of modern factorial research designs, requiring random assignment of participants to conditions and statistical inference based on null-hypothesis statistical tests (NHST) of group averages. By the mid 1950’s researchers in psychology were ‘addicted to p’ and the use of NHST became essential for research to be published. The second development occurring at almost the same time, was the development by the American Psychological Association of the scientist-practitioner model of clinical practice. This ideal rapidly became the dominant model for university training of clinical psychologists in the USA and has been generalised to the training of applied psychologists in general and across the world. Not surprisingly, the ‘scientist’ part of the scientist-practitioner ideal became closely associated with NHST-based research. Clinical and applied research has for nearly 50 years thus also been ‘addicted to p’, dominated by the search for statistical significance among group mean differences rather than clinical or practical significance and unable legitimately to make inferences about individual clients. The contemporary rise of the evidence-based practice movement, which can be considered a reformulation of the scientist-practitioner model, has brought sharply into focus again what has also been known for most of those 50 years: Our research methods, and especially our data analysis methods, are poorly adapted to the needs of practice. Research is about ideal, abstract, average types; practice is about individuals in all their diversity and variability. Furthermore, there is now an emergent ‘crisis’ in psychology due to the recognition that much of our research fails to replicate. I will review this lamentable history, and then consider some of the ways that we can adapt our research practices to make them much better adapted to evidence-based practice. These include the use of single—case research designs and novel methods of visual analysis of data. Reference: Blampied, N.M. (2013). Single-case research and the scientist-practitioner ideal in applied psychology. In G. Madden (Editor-in-chief). Handbook of Behavior Analysis Vol 1. (pp 177 – 197). Washington, DC: American Psychological Association.